# w-0161a

Toward a Stakeholder Theory of the Firm:

The Case of the Saturn Partnership

Thomas A. Kochan

Sloan School of Management

Massachusetts Institute of Technology

and

Saul Rubinstein

School of Management and Labor Relations

Rutgers University

May, 1997

Toward a Stakeholder Theory of the Firm:The Case of the Saturn Partnership

Abstract

This paper seeks to engage the organization theory community in contemporarydebates over the role of the cxxporation in American society by building a stakeholdertheory of the t%n. We develop a set of normative premises and positivist propositionsderived from these debates, from a combination of organization and industrial relationstheories, and by using data from the Saturn Corporation. Three questions are posed fora stakeholder theory: (1) Under what conditions is a stakeholder firm likely to emerge,(2) what are the critical determinants of performance in a stakeholder firm, and (3)what will determine the sustainability and diffusion of this organizational form in theAmerican environment? The history, design features, and dynamics of the labor-management partnership at Saturn are used to illustrate and interpret a specific case ofemployees as stakeholders. Saturn’s governance structure, work organization, andinternal processes fit the characteristics of a stakeholder firm. Employees establishthemselves as critical stakeholders by using their knowledge to improve organizationalperformance. The local union likewise contributes to firm performance by organizingworkers into a dense social network that contributes to problem solving, conflictresolution, and quality improvement. However, the legal and political environment inwhich the firm operates produces considerable uncertainty over the sustainability anddiffusion of Saturn’s features in particular, and the stakeholder organizational form ingeneral. Additional hypotheses and research questions are proposed to continue theorybuilding around the more general model of the stakeholder firm. Organization theoristsare encouraged to take up the analysis of stakeholder models and thereby contribute tothe contemporary and future debates over the role the corporation in American society.

Toward a Stakeholder Theory of the Firm:The Case of the Saturn Partnership

The purpose of the American corporation is to maximize shareholder wealth. From this

basic legal principle, flows the objective function and important features of the organization

desi~ control systems, governance stmctures, and the legal rules and norms that structure

employment relations in American firms. In recent years, however, both this conception of the

corporation and the employment relations system which it fosters have come under increasing

: as concerns are voiced over, among other things, corporate downsizing, stagnant real

wages, growing income disparities, increases in contingent work and declining union

representation (New York Times, 1996).

Some of those concerned about these issues implicitly or explicitly cast their critiques in

the language of multiple stakeholders. That is, they argue that firms are composed of multiple

interests or stakeholders each of which contributes to the firm and therefore has a legitimate claim

against its resources equivalent to the claims of shareholders (HuttoL 1995; Reich 1996).

Although the stakeholder idea has been advanced in the orgtitional theory literature for some

time (Freeman and Reed, 1983; Freemw 1984; Evan and Freemzq 1988; Donaldson and

lFunds for this research were provided by the National Science Foundatio~ the MIT InternationalMotor Vehicle Research Progrq and the MIT Leaders for Manufacturing Program. We wish tothank Peter Cappeu Adrienne Eaton Charles Heckscher, Larry Hunter, Paul Osterm~ JamesRebitzer, Sue Schu~ John VanMaaneL Karl Weiclq and F&ten Wever for helpfbl commentson earlier drafts.

Presto~ 1994; E~ 1978), it has yet to be defined in a way that aUows it to be developed into a

clear thecxy with propositions suitable for empirical testing. Organization theorists and

researchers will need to bring greater conceptual clarity and precision to this concept if the field is

to heed the call to contribute to theoretical and policy debates over the basic goals of the

corporation and its role in society (Stem and BarIey, 1996; Selznic~ 1996).

Aother set of critiques of the American corporate form has come from industrial relations

researchers and policy analysts who argue that the law and institutional arrangements structuring

traditional employment relationships no longer sene the needs of either workers or firms

(Kochq Ka@ and McKersie, 1986; Commission on the Future of Worker Management

‘ations. 1‘a~a~ 6 e school of thought in this field (Kochan et. al., 1986; Kochan and

mna I sts that the trial and error struggles to adapt traditional practices have

identified Ie ftires of a transformed model of employment relations that are better suited to the

current environment and needs of the parties. The transformed model seeks to increase flexibility

in the use of human resources and worker participation and influence at all levels of the enterprise,

from the workplace to the highest levels of strategic decision-making, in return for greater

commitment and contribution of the worldorce and their representatives to improve the economic

petiormance of the enterprise. It rejects a key premise built into American labor law and

shareholder mximizkg models, namely, that strategic decisions about how to compete or

produce can be separated fkom decisions over how to distribute the gains or losses that result

from organizational activities (Cease, 1937). In this way the workforce appears to become a

stakeholder in the h. Thus, the trantiormed industrial relations model might be viewed as a

specific case of a stakeholder @ one that focuses on employees as stakeholders.

Neither the stakeholder nor the transformed industrial relations perspectives have been

developed into a complete theory of the firm. Nor do we have well documented empirical studies

that explore the dynamics of organizations that approximate the models implied by these

arguments from which such a theory might be developed inductively. This paper attempts to

begin filling this void by using data born a case study of the Saturn Corporation to develop a

theory of the firm in which employees are critical stakeholders.

Our approach is to first present a set of normative premises and positivist propositions as

starting points for building a general stakeholder theory of the firm. These are derived from

contemporary debates about the corporation in society and ilom existing organization and

industrial rl s thee-” -- *-.-” s from our observations at Saturn. Then we apply these

general pro] nst< . ase of empIoyees as stakeholders and use them to interpret _

the experiences of mar ~ement and labor in implementing Saturn’s organizational principles and

design over the first decade of its life. The final section of the paper then returns to the more

general case and proposes additional hypotheses and research ideas for deepening stakeholder

theory. As such our approach fits Weick’s (1995) description of “theorizing” i.e., building theory

by iterating between inductive and deductive methods.

Saturn is an ideal laboratory for such an analytical effort since it appears to embody more

of the features of a stakeholder firm and a transformed employment relationship than perhaps any

other large scale organization in America today. Yet, as will be developed below, these features

also make it vulnerable to critics and opponents, making it also a study in the politics of creating

and sustaining a stakeholder firm in the American environment. Thus this case also oflkrs

insights into the broader political dynamics facing workers, unions, and others seeking to change

the balance of power and the role of the corporation in the American economy and society.

The Theoretical Task

The fist task of a stakeholder theory is one of definition. What defines a stakeholder h

and how does it differ from a conventional herican style profit maximizing firm? Unfortunately,

although the term is ofien used, it is seldom (if ever) explicitly defined. For our purposes,

however, we will define a siakehokier as an individual or group with an ongoing interest in the

economic well-being of an enterprise and that contributes valued resources needed for the firm to

be successfi.d. A stabholderjh-m then is one whose purpose is to achieve the joint objectives of

parties with different inter The s?z~-Wd- ~m seeks to maximize the joint welfare or utility

of the interests involved@ t995 lutethe returns produced equitably among

these interests. However, whether it - tiievesjoint maximization or requires some tradeoff

between or among the difYerentstakeholder interests is an empirical questio~ depending on the

firm’s performance. A key distinctio~ however, born a shareholder maximizing model is that

stakeholders have a legitimate claim over the residual resources of the firm. Implicit in this

definition is the notion that not all the stakeholders seek to “maximize” the same things. Owners,

for example will want to maximize the financial return on their investment. Employees may want

to balance their short term fiancial return (in the form of wages and benefits) against their

interests in long term employment security or their interest in fair treatment and

challenging/satis&ing work experiences. For a stakeholder firm to be viable overtime, it must

demonstrate its abtity to both achieve these multiple objectives and to distribute them in ways

that maintain the canmitment of the different parties.

4

A iiwther detitional clarification is required it as we do in this paper, focus on employees

as stakeholders. That is, how, if at all, is a stakeholder firm different from what is increasingly

described as a “high performance” firm (Lawler, 1986). Indeed, the two share many design

features. However, we see the key diiXerenceas follows. A stakeholder firm has a dz~erent set of

goals or objectivejiaction from a shareholder fum while a high petiorrnance firm follows means

or strategies to achieve the traditional goal of maximizing shareholder wealth that have the

additional benefit of producing positive results or mutual gains for employees. This is often

referred to as a high productivityfigh skills and wages strategy (Kochan and Ostennan 1994). -

But a high petiormance firm does not alter the basic objectives of the traditional firm.

This definition begs an importar stion: ~~ are tb- stakeholders? Oflen this term is

extended to include any group that cent $res- which the organization would ~

cease to exist (SU 1963) and thus employees, Supr ters, communities, even customers are often

included under this term. In this paper we take a more modest and limited approach and develop

a specific theory based only on employees as stakeholders and leave the development of a more

general theocy to a later date.

What questions should a stakeholder theory address? We bellieve the msot critical

question lies in explaining what would it take for stakeholder firms to emerge, peforrn effectively,

sutive and be sustained overtime, and increase in number? We believe the most critical question

lies in explaining why, if this is a desirable organizational forq don’t more of them exist in the

u.s.j ~ wh~ fouows, we bra these general questions down into the following discrete

‘acomponents: .

(1) What factors lead to the emergence or formation of a stakeholder model oforgankttion?

9

(2) What are the key organizational features (governance, desi~ internal processes,etc. that a&ect the performance of a stakeholder firm?

(3) What determines the sustainability and diflision of this organizational form in theAmerican environment?

The Normative Undeminnimzs of the Stakeholder Model

B&ore out!ining the positive elements of a stakeholder theo~ of the @ the normative

underpinnings of this conception of the firm needs to be made explicit. The soul of a good theory

!ies in its normative foundations. Freidman (1958) embeds classical and neo-classical economics

in the values of individualism and freedom. Weber emphasized the importance of the Protestant

Ethic in the rise of capitalism (Giddens, 1976). Marx’s explanations and predictions about the

evolution of society rested on the conviction that the’ ting capitalist svstI=mhe was obseiving

would lead to greater exploitation of the working clas. therti .i, .. ,.,tive economic -*

system in which the working class owned the means of production ,~ouldbe preferable.

Institutional theories evolved first in economics out of a conviction that the laws and research

methods of classical economics did not and should not apply to labor since labor embodies a

human

1980).

element that makes it more than just another “commodity” (Do- 1949: McNulty,

Yet despite the importance of normative assumptions and values in the social sciences,

organization theory in general, and theories of the firm in particular, have seldom examined the

normative foundations or values that influence the ihming of research problems and theorizing.2

22 The recent exchange on “what is not theory’ in ~~tive Science Ouarterly (Sutton and

Staw, 1995; Weic& 1995; DiMaggio, 1995) is an example. The values or the normative concernsthat motivate theorists and influence how they frame the questions of interest are not mentioned inthe various contributions devoted to theorizing in organization research.

6

Events, however, may force organization theorists to reexamine the normative underpinnings of

their theories and to reposition organization theo~ in its larger societal context (Stem and Barley,

1996). Indee& we maybe entering another historic phase over the debate of the American

corporation equivalent to those of the populists in the 1890s that gave rise to the anti-trust

legislation or of the New Dealers in the 1930s that gave rise to legislation governing financial

markets and corporate governance as well as the rights of workers to bargain with employers over

employment conditions.

Why are debates over the role of the corporation in society once again now emerging as a

significant political and analytical issue? One reason is that the pendulum maybe swinging back

after a decade of shareholder revolts that ushered in a renewal of in’ capita’:--- ~T---- ,

1996; Bowman and UseeW 1996). The net effects of the battle over mari -~‘- , ~1 :te h

control” (Jensen, 1989) of the 1980s and early 1990s has been to sh.iflpower tc ,nareholders,

perhaps at the expense of other organizational interests. Second, there is growing recognition of

a breakdown in the implicit “social contract” between employees and employers that seemed to

alIow firms and employees to jointly prosper in the three decades that followed World War II.

That contract essentially had two key elements: worker (especially manager) loyalty was

exchanged for long term security (Hecksher, 1995) and productivity and profits were exchanged

for real wage improvement (Tlwrow, 1996). The restructuring and downsizing of large firms and

increased willingness to lay off managers and other white collar employees broke the first element

of the contract. The stagnation of real wages from 1973 to the present and the growth in

inequalky of incomes broke the second element of the contract.

7

l%ir~ those who come from an industrial relations tradition are concerned about the

social and economic costs of the long term decline of the American labor movement. A deeply

held tenet of industrial relations theoV is that strong and independent institutions for worker

representation and voice are critical to a democratic society. Yet neither industrial relations

theorists nor policy makers are ind~erent to the forms worker representation and voice take in

the economy and at the workplace. Unions and other forms of voice can and do impose costs as

well as confer benefits on society and to the individuals they represent and the organizations in

which their members work. Thus, contemporary debates within the industrial relations profession

revolve around how to revive worker representation and the forms it might best take to meet the

needs of the workforce and perform its economic and democratic fimctions.

FinalIy, cument debates over the role of the corporation in society reflect a deep and

growing distrust of the public in a number of society’s institutions. Trust and respect for

corporations, unions, government, and religion have all declined consistently in the past decade

(Gallup, 1996). Relations across institutions—business-government business-labor, etc. appear

to have become more polarized (Kochan, 1995). Thus, the search for an alternative conception of

the firm should be seen as part of the broader reexamination of institutions in society that is likely

to dominate political and analytical debates for the foreseeable fiture.

Corporations, like any institutio~ must seine the interests of society. Thus, debates over

the appropriate purpose(s) of corporations ultimately comes down to the partly normative and

partly positive argument over which form (or under what conditions) better serves society’s

interests. The shareholder maximbin g model of the firm rests on the assumption that society’s

resources are allocated most efficiently through corporate forms that have one superordinate goal,

8

i.e., to maximize the wealth of those who provide it with financial capital. The stakeholder model

offers a different normative assumption: society is best served by corporate forms that seek to

organize its multiple interests in ways that maximize their joint welfwe and that divide these

returns equitably among the dfierent interests. Which of these conceptions is most appropriate is

partly a political question and partly an empirical issue Debates over both these issues will,

however, be great~ influenced by the performance of organizations that attempt to embody

stakehoIder principles in their design and operations. Thus we believe the major tasks for

organization theory and research are to develop the normative and positive elements of a

stakeholder theory of the firm, collect the empirical data needed to test these propositions, and

help develop the managerial methods and skills that help stakeholder firms prosper.

Frrwrnents of a Stakeholder Theow of the Firm

Fragments of theoretical arguments for a stakeholder theory can be found in disparate

literatures ranging from organizational theo~, the economics of organizations in general and in

particular those that compare American and Japanese corporations, and industrial relations,

particularly models of transformed labor-management relationships. Taken together, these

fragments suggest several tentative answers to the questions posed above. We will present these

features in propositional form to make them easily testable in fhture work and then illustrate them

with our data and experiences at Saturn. The reader should be aware, however, that as noted

above, in reality these propositions were derived iteratively as we sought to organize and make

theoretical sense out of our obsemations of Saturn over the course of this research.

The propositions presented below are first Ihrned in a general enough way to apply to any

potential stakeholder (e.g., suppliers, retailers, communities etc.).. From these general

9

statements, we then derive specific propositions for the special case of employees. We return to

the more general case in the Discussion section of the paper.

We organize the propositions around two axis that are critical to our definition of a

stakeholder and a stakeholder b. One axis addresses the characteristics and behaviors that

determine whether.or not an individual, group, or organization fits the definition and requirements

needed to be a stakzhokikr. The second axis addresses whether what we will call the governance

structure, i.e., the organizational and institutional features that together structure the production

and distributional fimctions of the enterprise, produce and, over time, reproduce, the stakeholder

firm.

Ouestion 1: When do Stakeholder Firms Emerge?

The shareholder maximhing firm emerged in American society when finance capital was S

the critical scarce resource or asset needed to form organizations of sufficient scale to take

advantage of mass markets (Roe, 1994). For other resources to gain stakeholder status they

must also supply critical resources or assets to the enterprise. Moreover, the relationship must be

reciprocal. That is, the success of the enterprise must be critical to the value of the asset so that

its owners can legitimately claim a “property right” equivalent to the property rights of the

financial investors who pooled and invested their resources to create and sustain the enterprise.

Finally, the contest for legitimacy and iniluence is not predetermined by natural forces; it is partly

volitional and strategic in the sense that not all participants want to commit their energies and

resources to the enterprise and share in governance responsibilities and not all have the power to

assert their c&@ to stakeholder status and intluence even if they have a Iegitiite claim to such

status. Thus, leadership, strategic choice, and power also influence the emergence of Stakeholder

●

10

firms. With these generic propositions in mind we can now explore the more specific conditions

under which employees might be expected to emerge as key stakeholders.

Erndovee Features

Knowledge As An Ass@. Why would we expect to see more organizational forms emerge

now and in the fb~re that embody employees as critical stakeholders? To see why this is the

case, we need to explore the role of knowledge as a strategic asset in the modern firm.

Firms rely on dtierent mixes of capital, material, technical, locational, and human assets to

achieve a sustainable competitive advantage. The modem American corporation and the

shareholder maximizing principle emerged in an era where access to large pools of finance capital

was the most important new resource to be mobilized (Berle and Means, 1932; Chandler, 1978;

Roe, 1994). More recently a variety of ahemative conceptions of the firm are evolving out of ~

models that view knowledge as a critical asset and source of competitive advantage (Drucker,

1980; Thurow, 1996; Arrow, 1996). The basic proposition in these models is that in

organizations where worker knowledge is a critical asset, human resource policies will be

designed to reduce the incentives or the ability of knowledge workers to leave and thereby take

the firm’s assets with them (Rebitzer and Taylor, 1995). One way to do so is to make employees

partners by giving them a stake in the financial petiormance of the firm and a voice in its

governance structures. In some cases this financial stake is explicit as in employee ownership

plans (ESOPS) (BIasi and Kru% 1991; Blasi, 1995). In other cases, pmicularly in professional

services such as law, health and consulting the orgtition may take the form of ptinerships

rather than public corporations. One study of law firms, for example examines the stnmtural and

incentive f=tures of pamerships in which employees can “hold up” the firm by virtue of their

11

ability to leave the organization and take their clients with them. This threat leads to alternative

organizational forms and compensation rules that bind partners to the organization by putting a

high percentage of their compensation at risk but rewarding partners well for their loyalty and

contributions to the firm (Rebitzer and Taylor, 1995). This perspective suggests that as

knowledge becomes a more critical asset or source of strategic advantage, organizations will

implement various types of explicit or implicit partnership arrangements, all of which serve to

treat knowledge workers as critical stakeholders. Thus, an initial requirement for employees to

become stakeholders is that the knowledge they possess is a critical asset to the firm.

Proposition 1:The more important employee knowledge is to the success of the fun, the more likelyemnl~’~eeswill emerge as an influential stakeholder.

Ie of SDecificHuman Caoital. Under what conditions would we expect employees ~

to ./ant or believe it appropriate to seek stakeholder rights in an organization? Blair’s (1995)

theoy of the role of human capital in firms provides some insight on this question. Blair (1995)

argues that employees share a residual risk in their employment relationship with the firm and,

therefore, should be residual ckhnants against the value created. Her argument focuses on the

specific human capital employees build up overtime in employment relationships and thus sewes

as the mirror image of the knowledge-as-an-asset proposition. She argues that the shareholder

maximizing model assumes employees are paid wages determined by competitive labor markets

while the providers of equity capital absorb the residual risks associated with their investments. If

all other parties to the organization also receive fixed payments for their contributions that are

independent of variations in firm specific performance, then mximizhg shareholder wealth is

12

equivalent to maximkhg the total wealth of the firm. But, since shareholders absorb the risk

they are entitled to control the corporation in their interests.

The argument that shareholders are the sole residual risk bearers breaks do~ however,

in organizations where employees either receive a wage premium for their specific human capital

and/or are compa,sated in part based on firm pefiormance. Where either of these conditions

hol~ employees bear residual risks similar to shareholders. In the former case, the risk lies in the

costs ofjob loss since the employees would not be able to filly replace their current wage in the

external labor market. In the latter case the risk lies in the variability of employee income that is

due to variations in firm petiommnce. Blair argues that in either case employees should be

en’ .0 a rol- !.. +4- - arnance and control of the firm. ~ both si~ations, B1~ (1995) makes

the ‘or n total wealth of the firm among all residual claimants (shareholders and ~

employees) ra+ .er than just shareholders. Moreover, we might expect that employees themselves

will want to achieve stakeholder rights and influence in settings where they are sharing residual

risks. One typically sees this interest expressed directly when employees purchase or negotiate an

ownership stake in their firm (BIasi and Kruse, 1991). However, if Blair’s arguments are correct,

a broader but more latent demand for the rights of ownership exist among employees that have

built up stocks of specific human capital in their tkms.

Proposition 2:Employees will be motivated to create a stakeholder firm to the extent that theirmmpemation is wntingent on firm pefiormance and/or their long run economic welfare isdependent on continued employment with the flnq i.e., when employees would experiencesignifwnt economic losses if required to find employment in the external labor market.

Governance Arranzement$

13

The Role of LeaderShin and Power. Strategic choice (Child, 1972), institutional

(Selznic~ 1949; Di.Maggio and Powell, 1983; DiMaggio, 1988), and resource dependency

(Pfeffer and SalanciIq 1978) perspectives on organizations all agree that leaders, i.e., those who

control the decisions over how to attract and maintain resources needed from the environment,

have some discretion over what weight to give to different environmental forces and yet are to

some extent constrained by the relative power of those forces. It is this combination of discretion

and constraints that gives rise to the notion that organizational decision-makers and leaders to

some extent enact their environments (Weiclq 1983) and, over time, shape the values or

organizational culture of the firm. This is patiicularly true of the founders and other leaders who

shape the firm in “ ‘vmative stages (Sche~ 1983). Thus, we see the initial choices as shaped by

a mixture of the 1 that i ! to different stakeholder interests and the power of the ~

difXerentstakehoiws to con. ain the choices available.

Much of the organizational behavior literature on stakeholders ascribes a great deal of

importance to the values and leadership styles of top executives. Freeman (1984) and Donaldson

and Preston (1995), for example, argue that managers must view their roles as one of balancing,

coordinating and integrating the interests of the firm’s multiple stakeholders. This role definition

however, departs horn the more widely accepted defition of management’s role as an agent of

shareholders’ interests.

Organizational researchers who have employed this perspective in empirical studies

(Kotter and Hesk~ 1993) also emphasize the leadership style of top management. They argue

that some CEOS’ vision or personal values will lead them to adopt a stakeholder perspective in

dealing with employees, shareholders, customers, etc. Regardless of the legal or organizational

14

stnactures under which thq worlq these leaders see their role as one of balancing the needs of

multiple constituencies and thus shape their behavior and the policies of the firm accordingly.

Little attention is giv~ however, to why some CEOS would adopt this perspective, or what

structures or incentives determine whether or not this leadership style is supported or

institutionalized in-on-going organizational operations and through leadership successions. Thus,

while hardly a sufficient conditio~ this view suggests the following proposition:

Proposition 3:Top executives must embody the values and leadership styles consistent with a stakeholderperspective for this organizational form to emerge.

Industrial relations theoiy shares the view that organizations have some discretion in

choosing how to respond to o~ ‘ their - “--m-- ;, however, industrial relations models

tend to give greater weight to F relat” ~ ; the contending interests or external --

resources (Koch~ Kaz and McKersie, 1‘ 36). The industrial relations view goes roughly as

follows. Leadership is unlikely to be enough. Strong forces will resist the sharing of power and

returns with non-shareholder interests. Some independent source of power is likely to be needed

for employees to gain sufficient influence in decision-making to create a stakeholder firm. Unions

are the tradltiona.1source of independent power for workers. However, traditional unions

embedded in traditional bargaining relationships are kept at arms-length from management and do

not have access to the strategic decisions afFectingorganizational design. Instead, unions (and

workers) are, acwrding to both American law and custo~ supposed to focus only on

distributional issue% leaving strategic issues to management. Thus, for employees to have a voice

in these decisions requires a union leadership that accepts and champions the features of a

tran@ormed employment relationship (Kochq Kaw and McKersie, 1986).

15

The theory of trantiormed industrial relations was derived from the problems observed in

the 1980s with the New Deal industrial re!ations system (Kochan, McKersie, and Cappelli, 1984;

Ka@ 1985; Kochzq K@ and McKersie, 1986). The essence of the New Deal model is that (1)

managers, as agents of the shareholders retain the rights to make entrepreneurial decisions fkee of

interference of workers or their union representatives, (2) workers and their unions have the right

to negotiate only over the impacts of managerial decisions on wages, hours, and working

conditions, and (3) the employment relationship on a daily basis is regulated by the terms of the

labor contract and workers achieve a voice through their union representatives or by filing a

~“evance when they believe their rights under the contract have been violated. Over time, this

system leads to increasingly detailed written qovernino the lah~r process.

The transformed model grew out of c nent. nions, and companies

improve peflormance by introducing greater nexibility, ~ .,ployee participatio~ and union-

management consultation into decision-making. A central proposition in this model is that efforts

to transform the relationship at one level of the firm depend on achieving reinforcing changes at

other levels. That is, employee participation at the workplace will only be sustained over an

extended period of time if reinforced by employment security protections or assurances, and an

effective voice for worker interests in the strategic decisions that affect the long run future of the

enterprise. ‘I%M the transformed model of industrial relations envisions a firm in which

employees eum a voice in the governance process by contributing to its economic performance

through their participation in decision-making and by increasing flexibility by giving up some of

the work rubs that protected them in the traditional New Deal model.●

16

Thus, we suggest a second feature of the governance structure or process needed for a

stakeholder fkm to emerge in the American environment:

Proposition 4:

A stalceholder design for an organization is more likely to be chosen when employees havea union or other independent source of power whose leaders believe in taking an activerole in the $rategic management of the firm.

Question 2: What are the Critical Features needed to make a Stakeholder Firm Succeed?

A stakeholder firm is composed o~ legitimates, and gives voice to the existence of

multiple interests each of which has goals that must be satisfied. Since the activities of the

stakeholders are highly interdependent, i.e, the behavior of one affects the behavior of others,

their interests can be both shared and different, and powe: z Wi( ! across--

groups and throughout the hierarchy, there is a high potenti Oo:t. JndkL ~wmmidt and

Koch~ 1972; Je~ 1995) and for the enhancements to performance that potentially resides in

groups or teams (Cohen and Bailey, 1996) . Thus, the critical organizational tasks for a

stakeholder firm are to (1) mobilii the stakeholders to commit their assets in ways that

contribute to performance and (2) resolve the conflicts that arise when multiple interests share

power in the governance process. These requirements have profound implications for the

psychological or social contract between the employee and the firm and for the structures and

processes that govern employment relations.

Emtiowe Features

Use of Discretionary Knowledge and EfforL For employees to be stakeholders they must

contribute to the success of the firm by using their discretionary effort (i.e., their knowledge or

17

other sources of human capital) to enhance firm petiormance. This can take the form of

improving productivity and quality through continuous improvement efforts, contributing to the

innovation process by creating new products or by reducing the time required to bring new

products to the market etc. The key feature, however, is that employees remain motivated and

committed to contributing to firm performance and do not fkIl into behavior patterns described in

the classic studies of group norms (Chinoy, 1955; Roy, 1953) which produce standard or

accepted production rates that are enforced by group norms and in which “rate busters” are

penaliied. The organizational governance features obviously play a role in shaping these norms

and will be discussed below, however, collective and individual employee attitudes and behavior

also influence effort norms and behavior. oThis is why firms that seek to z A“high

commitment” norms (Walto~ 1985; Lawler, 1986) often prefer to build t . .. rgan?, . ‘..,,s,, ---

greenfleld sites in which employees can be carefidly selected, screened, ana socialiiej .O fit the

high commitment model of effoti and behavior, recognizing it is more diflicult to modifi work

and group norms once they are established in an organization’s work culture. Whether or not a

new site and carefidly selected and socialized worldorce is available or the workforce comes with

an embedded work culture or no~ employees must be willing and able to use their discretionary

knowledge and effort to contribute to enhancing firm performance in significant ways for them to

behave as stakeholders.

Proposition 5: Employees must use their discretionary knowledge and effort to contribute to firmpdorrnance if they are to be stakeholders in the firm.

Governance Featur~

Aoki’s J Form. A ckar conception of the features of the stmture and employment

practices practices of a Japanese stakeholder firm is found in Aoki’s (1986; 1990) comparison of

Japanese and American corporations. Aoki describes the Japanese corporation (J-form) as a

coalition between shareholders and employees with managers sewing as mediating agents. The

dense network of cross-holding among shareholders (largely institutional investors such as banks,.

insurance companies, and other companies linked together within a common group or keiretsu)

provides stability and patient capital. Shareholders receive their returns through a combination of

long term capital gains, interest on debt securities, and income horn direct sales to the firm.

Emp[oyees are rewarded with employment security, promotional oppofiunities, wages tied to

seniority and performance, and retirement benefits that reward long service. Sharehol ~nd

employees have a common interest in organizational growth and stability. Human rest. . .1 *

practices support pursuit of these shared objectives through heavy investment in on-the-job

training, employee suggestion systems and quality circles, and extensive labor-management

consultation. Human resources executives are paid salaries equivalent to other executives and

play an influential role in the top governance structures of the firm.

Information exchange and coordination play key roles in Aoki’s model of the Japanese

firm. Horizontal coordination that occurs through information exchanges substitutes for the

hierarchical control systems built into the traditional American corporation (Aoki, 1990).

Horizontal coordinationand ifiormation sharing are encouraged by organizational design fatures

that break I%omthe American legal and managerial traditions of separating those who plan and

supervise from those who execute work assignments. Among these features are job rotatio~

enterprise unions that include blue collar, white coUar, and managerial employees, extensive use of

19

teams and other problem solving processes that support the sharing of knowledge needed to

improve work unit and organizational performance.

While Aoki developed his conception of the Ilrm based on a simplified prototypical

Japanese organiz40~ he suggests that it is not culturally bound to Japanese society. Rather, it

may fit those environments in which markets demand product variety, where the technology can

be Mormed by and made more productive with worker knowledge, where the market conditions

support a premium on qualhy, innovation, and rapid response to changing consumer preferences,

and where the regulatory environment supports these organizational featmes.

Thus, unlike the American organization theorists who argue top management will be the

catalyst for implementing a stakeholder perspective by shaping the organization’s culture, the

Japanese corporation’s structure, financial markets, governance arrangements, coordination.Xmli=

processe$ and human resource practices serve as complementarities (IW@romand Roberts, 1993)

that give rise to and institutionalize this conception of the firm. This view therefore suggests that,

to function effectively, organizations that treat employees as influential stakeholders will need to

have the following complements f~tures:

Propositions 6a-e:a. Sources of capital that are patient i.e., that focus on long run returns to

shareholders, and that derive their returns in multiple ways;b. Governance structures that provide an influential role to employee interests and

human resource considerations;c. Coordination mechanisms that achieve high levels of vefiical and horizontal

tiormation exchange and that decentralize planning and supewisozy fimctions tothose doing the WO~

d. Human resource polici~ e.g., compensatio~ training stafling and employmentsecurity, etc. that build loyalty and commitment to the firm.

e. Labor-management ktitutions that are enterprise based.

20

The Role of Confkt and its Resolution. Conflict theo~ (Schmidt and Kocha~ 1972)

would predict stakeholder firms would experience high levels of conflict because of the increased

influence of employees that had heretofore been in a low power position and the increased

interdependence among parties in a mixed motive setting. Conflicts that in the past were

suppressed by either an imbalance of power or by iimiting interactions among these parties

through hierarchical or fictional specialization can now be more readily surfaced. These conflicts

can either be constructive in that they identifj individual concerns and organizational problems

that need to be addressed in order to maintain commitment and improve performance, or if not

resolved satisfactorily, can generate conflict traps and produce a recurring cycle of low trust and

high conflict (Fo% 1974). This implies that conflict resolution becomes a more important

organizational process in a Stakeholder firm. On the other hand, the changed ‘incentivestructuresm

of stakeholder firms, i.e., the acceptance of the legitimacy of and the shared burden of pursuing

multiple interests, might imply that mutual gains are possible if the parties are willing and able to

adopt integrative bargaining strategies and behaviors (Walton and McKersie, 1965; Walto~

McKersie, and Cutcher-Gershenfeld, 1995). In either even~ regardless of whether one predicts

greater or less potential confli~ the critical task for the organizational and industrial relations

processes in which contlicts surfkce and are managed, therefore, is to promote the effective

resolution of conflicts in ways that enhance the achievement of both separate and shared

objectives (Cutcher-Gershenfel~ 1990 ). However, ineffective cordlkt resolution that

degenerates into a low trusthigh conflict cycle or equilibrium will exact a high price on the

petiormance of a Stakeholder firm (Jem 1995). For this reaso~ we see cdlict resolution

processes as a critictd feature of the governance stnwture of a stakeholder firm. The governance

21

structure must give voice and representation to employee interests in both production and

distributional decisions, but in doing so conflicts must be resolved in ways that both add value to

the firm and address critical stakeholder concerns and interests. Thus, the transformed perspective

on industrial relations continues to emphasize the value of collective representation of workers by

a union or some other institutional arrangement, however, the union or alternative institutional

arrangement must fimction in away that aids articulation and aggregation of employee interests

and conflict resolutions that add value to the enterprise.

Thus the role of a union or other collective body designed to give voice to workers as

stakeholders must change substantially in a transformed employment relationship. Under the New

Deal model, unions existed as a countervailing force whose power was derived from their ability

to withhold their members’ collective labor (Gailbrait~ 1967). If unions contributed to improved -~

productivity or efficiency it was only indirectly through their “shock effects” on management

(Slichter, Healy, and Livemas~ 1960; Freeman and Medofi 1984). That is, as unions increased

wages, benefits, and working conditions, management had to find ways to become more efficient

in order to recoup the costs of these improved benefits. While periodically unions and employers

would seek to cooperate and to improve their relationships, these episodes seldom survived over

time and were not institutionalized into the structure and process of management or the

governance of the h (Gershenfeld, 1987). Instead, unions came to be viewed as outside forces,

or third parti~ only to be brought in by workers when they deeply distrusted management.

.The transfomwd model of industrial relations radically alters the traditional view and role

of unions, o;o~her forms of collective representation. While unions continue to be held

accountable for representing their members’ economic and other interests as employees, they are

●

22

also responsible for adding value to the enterprise by facilitating employee participation and voice

throughout the difiTerentlevels of the organization. The old adage associated with traditional

industrial relations that “management manages and workers and their unions grieve is replaced by

a perspective that treats worker representatives as active participants in the management and

governance process. In return for a voice in management, union leaders are expected to accept

greater responsibility for organizational decisions. The role of union leaders thus becomes one of

both representing members’ interests in negotiations and facilitating worker involvement and

voice in decision-making on a more continuous basis. This becomes a diflicult balancing act, one

that few union leaders have been successful in carqing out over extended periods of time. Thus,

union leaders can expect to encounter significant political risks in adapting to their new roles in

holder firms.

Proposition 7a-b:a. Collective representation of workers contributes to the effectiveness of a

stakeholder firm by surfacing and resolving contlicts and facilitating workerinvolvement and voice in problem solving, decision-making and administration ona continuous basis.

b. The form of representation adopted must achieve an effective balance betweenadding value to the enterprise and representing the specific interests of thewortiorce.

Question 3: Are Stakeholder Firms Sustainable in America?

Corporations are assumed to be going concerns. That is, they are expected to have an

indefinite life, one that goes beyond its founders or original participants. Legal rights, contracts,

and procedures govern the transfer of ownership and responsibtities assumed by new entrants to

the organization. However, firms can and do fki.1,the importance of difhrent assets change over

23

time, and new leaders take up positions of influence who may not share the same values and

commitments of their predecessors. Moreover, any single enterprise is embedded in a larger

normative and institutional setting where ideology and interests compete for legitimacy and

resources, and are reitiorced by legal doctrines and requirements that have been enacted in

response to and retiorce the prevailing norms, ideologies, and distribution of power. Any change

in the Iegal status of stakeholders in the corporation will therefore likely provoke a political battle

since it will challenge the prevailing ideologies regarding property rights, governance rules and the

Iaws regulating the roles of difTerentorganizational participants. Thus, for a stakeholder

conception of the corporation to survive over time and diffise to significant numbers of

enterprises, it will need to not only demonstrate its ability to petiorm well, but also maintain the

con-’ tcessive generations of participants and leaders and manage the political battles ~

needt to change prevailing ideologies, laws, and norms. As we will see, this is particularly

problematic for employees as stakeholders because stakeholder firms cotiont not only the norms

and ideologies and power of corporations but also deeply embedded ideologies and norms

regarding the role of employees as collective interests, i.e., the role of unions in American society.

Emplovee Feature$

One frequently obsemd f=ture of employee owned iirms or partnerships is that they tend

to dissolve when the original owners or partners turnover, retire, or profit sufficiently from their

investment that they are induced to sell their “shares” and allow the firm to be transformed into

traditional ownership and governance arrangements. At a more psychological level, later

generations of new hires may lack the same high commitment and willingness to use their

discretionary tiosts to enhance firm performance as those present in the early years of the

24

organization’s life. Similarly, to the extent that stakeholder perspectives are not shared by all

who come into leadership positions in organizations, stakeholder firms are vulnerable to

leadership succession episodes. Thus, how to sustain the economic and psychological

commitment of successive generations of employees and leaders to their stakeholder roles appears

to be a serious challenge in making the stakeholder firma going concern. Indeed, this maybe.

one of the reasons why stakeholder firms are difficult to sustain in an American environment

where labor mobility tends to be relatively high by international standards. To overcome these

tendencies there needs to be some broad recognition of the value created by the stakeholder

model and incentives for successive generations of organizational participants to maintain and/or

develop ~eattributes through socialization.

Proposii . F02 %ms to survive overtime there must be means of transferring . zbot .-~.. #cd and economic stakeholder attributes across generations oflead, sand employees.

Governance Features

The Political Environment of Stakeholder Firms. Stakeholder models have come and gone

at various points in American histo~. In the 1960s, for example, a debate arose over the social

responsibilities of corporations. Advocates argued that corporations should voluntarily allocate

some of the firm’s resources to address social objectives believed to be important to society. Yet

these aqyments have been severely criticized by defenders of the shareholder maximizing model,

ranging from Milton Friedman (1970) to the editors of The Economist (1996), to those in

powerful finance and economic policy making positions in business and government (Business

Week 1996). While the social responsiii argument is not pafkctly analogous to the

stakeholder conception developed here, we take away from this histoIYthe expectation that

2s

arguments for an alternative to the shareholder maximizing model are likely to encounter a hostile

response from the mainstream finance and economics profession and the dominant political

culture. & such organizations that implement a stakeholder model represent a political

intemention as well as a economic and organizational experiment.

Proposition 9:Firms that “seekto institutionalize stakeholder principles into their operations will facestrong opposition flom traditional allies of the shareholder maximizing model in dominantpositions in business, govemmen~ medi~ and academia.

The Politics of Transformed Industrial Relations. Like the stakeholder conception of the

firm, the transformed model of industrial relations lives in a highly charged and rather hostile

political and legal em” nent. l%t the nr~’‘ailinglabor law limits it by drawing a clear line of

demarcation between lhts . ‘ ities of “employees” and “managers” as both

individuals and collective groups ~ ieiler, 1990; Gould, 1994). Second, a political stalemate

between business and labor over how to change labor law has prevailed for the last twenty years

(Mills, 1978; Koch~ 1995) making it difficuh to update the law to accommodate the worker and

union roles required for a transformed employment relationship and stakeholder firm to fimction

effectively.3 Third, there is a major debate within the labor movement over the wisdom and

3The clearest example of the stalemate is the fhilure of the Commission on the Future of WorkerManagement Relations created by the Clinton Adminktration in 1993. That Commission wasasked to recommend how to modernize labor policies to allow for greater employee participationand to Overmllw the high levels of conflict assockd with the union orgdzing and recognitionprocess. Members of the Commission differed over how far to go in recommending changes in thelaw that would support the transformed model of industrial relations and for di.ffkremtreasons, bothorpized labor and business opposed radical changes to the traditional system. In the end, thisopposition resulted in a set of compromise recommendations (Commission on the Future of WorkerManagement Relations, 1994b; Kochan, 1995). Even these, however, were too controversial forboth labor and managermt and the political stalemate continues.

26

viability of the types of partnerships embodied in the transformed model. This debate is

ptiiculady strong within the United Automobile Workers (UAW) (Parker and Slaughter, 1988).

Fourt& there is strong ideological opposition within business to the features of the transformed

model that increase !abor’s voice and power in strategic and governance issues and processes

within the corporation.

Thus just as the stakeholder conception of the firm operates in a highly charged and

hostile political environment so do efforts tot- much less institutionalize, the features of the

transformed industrial relations model. Stakeholder models that embody transformed industrial

relations stn.wtures and process= therefore, can be expected to engender considerable resistance

and debate.

Proposition 10:Organizations with transfom Mus.., relations wdl encounter considerable resistanceand debate since some of their features - dy violate prevailing labor laws and challengedeeply held ideological positions regarding the role of unions.

Methods and Data

The data for this study come flom a long term research project with the Saturn

Corporation and UAW Local Union 1853. Since Saturn’s inception we, along with other

members of our research group, have tracked its history through discussions with the union leader

who championed the project within the international union and with his counterpart in the General

Motors Corporatio~ plant visits during the years prior to the production of the first car in 1990,

and since thq through on-site interviews, observations, surveys, and participation in meetings

and seminws with union and management officials. Our relationship with Saturn reflects the

partnership norms embedded in the organidon’s mission statement and core principles. From

2’7

the begixudn~ union and management representatives urged us to work with them as partners

rather than “study” them at a distance as outside researchers. This provided us with excellent

access and cooperation. At the same time, our relationship was and continues to be one of

researchers, not consultants to the firm or the union. We have, however, fd back results of our

work to Saturn and UAW representatives earlier than would be normal for outside social science

researchers. In several cases, our results led to changes in the partnership designed to address

problems we obsewed. The data for this study therefore consist of a mix of historical

interpretation based on intemiews with the founders of Satu~ direct participant observation and

intention in the partnership, interview and field notes, internal survey data collected by the

local union and/or the company’s organizational ( oment ------- ~.J Umey data collected =

part of our own research program. Together the c -jlle~l; * :-. ,ISthe period horn e

1990 to January, 1997.

We present our interpretations of events at Saturn as an exercise in developing a grounded

theo~ (Gkser and Strauss, 1967) of the stakeholder perspective on the firm. We did not go into

this project with a stakeholder model in mind. Instead, the idea of using a stakeholder perspective

to interpret what we observed at Saturn developed as we searched to find a way to conceptualize

and assess the dynamics of the labor-management partnership that is embodied in Saturn’s

orgtitionrd design and is evolving overtime. In tu~ the more we explored the nascent

literature on stakeholdq the more it appeared that Saturn fit many of the f=tures discussed in

this literature. Thusj we saw an opportunity to both interpret what we were obseming and

provide em&i@ data needed to fbrther develop this model of the firm.

28

In what follows we will describe the history, design features, and dynamics of the Saturn

partnership in ways that both illustrate (testis too strong a term given the iterative way in which

the theory was developed) the propositions summarized above in practice and demonstrate the

diflkulties encountered in creating and sustaining this type of enterprise in the American

environment. While we present this as a single case study, wherever possible, we use variations in

practices obsemd across different individuals and organizational units and over time at Saturn to

provide a more conventional test of some of these propositions (Harris and SuttoZ 1986).

Specifically, we report the results of a network analysis designed to test whether and how one

unique f-ture of the partnership, the co-management process at the work unit level, contributes

to enterprise petiorrnance.

Results e

Saturn’s Creation: The Role of Leadership and Power

Table 1 presents a timeline of critical events in Saturn’s history to date. Saturn is a wholly

owned division of General Motors (GM) with manufacturing operations in Spring Hill, Tennessee

that produces sedans, coupes, and station wagons for the small car market primarily for sale in

North America. It currently employees approximately 9,000 workers, approximately 7,300 of

which are members of the UAW. Production began in 1990. In 1995 production was

approximately 290,000 vehicles and in 1996 production expanded to approximately 310,000

vehicles.

In 1982 GMs engineering staff conducted a study that concluded under existing

management practices and labor-management relations the company could not profitably build

small cars in North hwri% and therefore GM formed a joint alliance with Japanese and Korean

29

producers to import cars for this market segment. GM estimated that overseas production of

small cars could save “$2000per vehicle over its domestic production costs. Concern over the loss

of jobs in the U.S. led AMed Warre~ the Vice President of Labor Relations at GM and Donald

Ep~ the Vice President of the UAW for the GM Department to propose formation of a joint

study team to explore whether it would be possible to design an organization and employment.

system capable of producing a small car competitively in the U.S. with American workers under a

UAW contract. What became known as the “Committee of 99” ultimately recommended a

radically new organizational form in which work would be organized into teams, work rules

would be drastically simplified, and the union would be a partner in decision-making from the

bottom to the top of the organization (0’Toole, 1996). Saturn was also tos ~ a ]abo--+n~

for imovation that might be transferred back to GM. The preamble to Saturr - L. gin~ ~~+‘ - -

collective bargaining agreement included:

Saturn and the Union, as early as the spring of 1983, recognized the needfor a new approach tounion/management relations and the more eflective use of human resources t~assembt’y of smallcars was to be fem”ble in the United States. GMand the Vhion met and authorized theestablishment of a study center to explore the possibility of innovative approaches to staflng andoperating a manufactun”ng and assembly operation for the production of Smal! cars in theUnited States.

Further, Saturn’s Mission Statement has two parts:I. To markzt vehic!es dweiopedand manufactured in the United States that are worldleaders in quality, cost and customer sati$izction through the integration ofpeopIe, technology,and business systems.2. To tr@er knowledge, technolom and experience throughout General Motors.

In a 1991 intewiew, Roger Smi~ GM’s CEO during the creation and early development of

Saturn stated its goal axImproving the eflciency andcomptitiveness of every plant we operate... Saturn is the byto GM’s long-term competitiveness, surviwd, and success as a domestic producer.

30

Thus, from its out- Saturn was created to address the “interestsof shareholders for a small car

that would be profitable and help the rest of GM to learn from this experience and the interests of

GM employees and the UAW to provide jobs for UAW members.

These were controversial and ns~ proposals for UAW leaders. There was considerable

opposition within the UAW Executive Board since these ideas deviated born traditional seniority,.

compensatio~ and work organization principles and took the union into the management and

governance process in ways that traditiontdkts felt would jeopardue the union’s independence.

Clearly, without the strong personal leadership and support of Ephli~ the UAW would not have

ratified the Saturn labor agreement and gone along with the new role for the union (0’Toole,

1996).

The proposed design for Saturn provoked equally strong debate and controversy w

GM and in the broader business community. One of the most controversial features of the p

was that the company would voluntarily recognize the UAW in the new facility before any

workers were hired. This provision was later challenged by the National Right to Work

Committee, national anti-union lobbying group, on the grounds that such pre-hire agreements

violate the National Labor Relations Act. The case was eventually dismissed by the National

Labor Relations Board in 1985 shortly after President Reagan visited the new Saturn facili~ and

described the union-management agreement as the most important experiment in labor relations in

the country.

This early history is consistent with Propositions 3 and 4. Leadership was important to

creation of this stakeholder mode~ however, in this case leadership within the union was equally

important to leadership within the company. Moreover, without the presence of a strong union

31

able to assert its voice in the strategic decisions that had to be made to embark on and approve

this proj~ the stakeholder design would not have been chosen. GM would have continued to

import small cars from its international partners. But, as will be noted below, the internal

opposition within both the union and company (and the larger business and labor communities)

did not go away aqd would continue to af%ct the evolution and sustainability of this enterprise.

Ormnizational Desizn and Emrdovment Practice$

The fictional equivalents of many of the structural features of the stakeholder model as

described by Aoki were built into the organizational design and employment system of Saturn.

Table 2 summarizes these features.

Worker knowledge and commitment to improvement are seen as key organizational

attributes for accomplishing Saturn’s objectives. To achieve high levels of skill development new q

Saturn members received from 350 to 700 hours of training before they were allowed to build a

car. Recognizing the need for commitment to continuous organizational learning, the union in

1991 proposed linking training to the yearly risk and reward compensation plan. Eve~ Saturn

employee’s pay is tied to an annual organization-wide goal of obtaining at least ninety-two hours

of additional training.

32

Work is organized into self managed teams (called work units modules), 80 percent of the

workforce is covered by a no layoff commitment, minimal compensation is tied to the industry

average, however, additional rewards are contingent on meeting negotiated pefiormance criteria,

decision-making is based on consensus principles operating through “decision rings” at the

business unit (plant), manufkturing council (manufacturing operations), and strategic action

council (senior executive) levels, and information is to be shared openly and widely throughout

the organization. The local union is represented in each of these forums as well, @as will be

described below, in the line management structure.

The features of the organization design are quite consistent with those identified in Aoki’s

J-form firm and organizations (propositions 6a-e) and with the view that worker knowledge is a

critical organizatiomd asset (jmoposition 1). The risk-reward compensation syste~ along with ~.

the high base wages and benefits of autoworkers make employees residual risk holders at Saturn.

They would experience substantial economic losses if this organization failed. This motivates

employees to assert their interests as stakeholders in the management and governance processes at

Saturn. The next logical questio~ therefore, is whether these structural fatures, i.e., horizontal

coordination through infomnation exchange, joint governance arrangements, contingent

compensatio~ job rotatio~ strong enterprise-focused local unions, etc., have been translated into

worker behavior and organizational processes that in fact contribute to enterprise peflormance.

Interest in this question led us to focus on the co-management process in action.

The Co-Manamxn ent Proc~

Aoki’s model stresses the importance of lateral communication and “tiormation sharing as

a means for coordination and control in a stakeholder firm. At Saturq the co-management

33

process, and particularly, the role that union partners play in the management process, appeared

to us to be an example of this in action. Indeed, the local union’s internal structure and

governance process created a large and dense social network of representatives located in key

decision-making positions across the work units and the three plants in the Saturn manufacturing

complex. In total, over 400 members serve as partners to a manager and/or as an elected union.

officer at Saturn. The local union organized a large number of opportunities for these members to

interact that went well beyond the sparsely attended local union meetings typical of most local

unions in the U. S. These included such things as hi-weekly “Congresses” in which all officers and

appointed partners were expected to attend and at which a mixture of local union affairs and

issues related to the partnership and operations at Saturn were discussed, “one on one” member

surveys in which these officials interviewed local union members about problems or issues at

Satu~ social activities such as softball and bowling teams, etc. We also observed a number of

incidents in which these local union members and officers would draw on their networks to solve

problems informally without going through normal hierarchical channels. In one case during an

intemiew we were conducting with module advisors in the Panel Assembly module, a team leader

came into the office stating a team member thought the torque on his air gun was improper for

screwing down the panel under the wiper blades. Improper torque would lead to a loose panel and

possibly some rattling. The module advisor called down the line to the Car Final module and

asked his counterpart there to check the wiper panels. Upon learning that the defect had been

passed on and was likely to be present in cars ready for shipping this module advisor mobilized

several colleagues to repair all defective vehicles including those in possession of the carrier

(against company security rules) before any cars with loose panels were shipped. This rapid

34

mutual adjustment between module advisors in different departments was impressive. Equally

surprising was their willingness to take risks and responsibili~ for fixing problems rather than

sending them up the chain of command for a decision. This indicated a certain level of trust, and

strong horizontal communication and coordination between co-managers. Observation of

examples like this led us to hypothesize that the co-management process was behaving like a

dense social network of lateral contacts that contributed to itiormal, on-line problem solving at

Saturn. To test whether this was the case, we carried out a network analysis of the co-

management process among the work unit module advisors, the rough equivalent of a second

level of supervision in a traditionally structured firm. The detailed results of this analysis are

presented elsewhere (Rubinstei~ 1997) and are summarized in Figure 1 and Table 3.

specifically, we tested whether the communications occurring within this network of represented

and non-represented managers were related to quality petiorrnance.

The Communications and Coordination Network

Represented and non-represented module advisors completed daily communications and

time use logs for a two week period in May, 1993 to provide the data on the frequency and

content of their communications with other module advisors and the technical and fi.mctional staff

supporting the manu&cturing process (e.g., engineai.ngjmaintainence, etc.). These data were

subjected to a network analysis that described the density of communications, the centrality of the

represented and non-represented module advisors. and the specific topics dkcussed, and then

related these communications patterns to the quality performance (levels of first time quality and

degree of improvement in first time quality) for each module.

Figu~~ 1 contains a graph of the communications network of the module advisors. It is

partitioned intcthe three business units (plants) with Vehicle Systems (assembly) in the upper left,

Body Systems in the center, and Powertrain (engine and transmission manuticturing) in the lower

●

35

right. Each plant is fimther partitioned between union and non-represented module advisors. For

purposes of comparison communications densities within each group of union and non-

represented module advisors for each business unit are reported. The numbers reported in Figure

1 indicate that during this two week period the union represented module advisors in vehicle

systems communicated regularly with 26.3°/0of the other module advisors in this business unit.

This compares to i 7.3% of the non-represented module advisors in vehicle systems who

communicated with each other over this time period. In both Vehicle Systems and Body Systems

the union module advisors have higher communication densities than do their non-represented

counterparts. Powefirain densities for both groups were the same.

These communications densities were firther broken down according to the topics

disc The “’ dity (number of links) and density measures as well as

com .tior 1 sues were then related to the quality levels and rates of

improvements L each module. These results are shown in Table 3. The modules were divided

into two groups - High Fkst Time Quality Improvement and Low First Time Quality

Improvement. As we can see from Table 3, the grouping of modules with the highest level of

quality improvement also had significantly higher levels of communications by the represented

module advisor. This was true for communications centrality (the overall level of

communications), and for group centralky (communications with other represented module

advisors within each plant). Most striking are the differences in communications frequency,

specifically on the subject of quality.

Departments were also divided into two groups based on their 1993 level of first time

qual@. As Table 3 shows, the level of communication and coordination is related to fist time

36

quality performance. The overall site-wide communications centrality of the represented module

advisors is significantly higher in the High FTQ group. Similarly, the density of communications

among represented module advisors within each group (plant) was significantly higher in the High

FTQ group. Communications on quality was significantly higher for both represented and non-

represented module advisors in the High FTQ group, although the represented level of 3.46

quality communications was greater than the non-represented level of 2.21. Regression results

controlling for differences in technology and business unit confirmed the significance of the

relationship between communications density of the represented module advisors and quality

performance (Rubinstei~ 1997).

In summat=- ‘be network analysis shows that(1) union module advisors participate in a

dense communica Letwe oeers and with their management appointed

counterparts, and , .Ie nwr such communications occurs, the higher the quality performance

of their module. Thus, where the co-management process functions in this fh.shion,the union-

management partnership is adding value to the corporation in a fashion consistent with that

predicted by Aoki’s model of a stakeholder firm and proposition 7a pertaining to the role of a

union in a transformed relationship and organization.

37

The Union as a Remesentative Body

Propositions 7b suggests that a union in a stakeholder firm must also continue to perform

as an effective representative of its members’ specific interests. How to balance this role with its

role as a partner in management has been a difficult challenge for the local union at Satu~ as it

has been historically for American unions engaged in other types of union-management

cooperative relationships (Gershenfeld, 1987). Some of the diilicuhy may relate to the

complicated local-national union stmctures of American unions, however, some may reflect the

inherent tension between intra-organizational relations in a democratic union engaged in mixed

motive relations with an employer (Walton and McKersie, 1965).

Aoki lists an enterprise u I as one of the complementary features of the J-organization,

American unions such as the ont :senti ‘ Saturn have several features that

diiTerentiate them from enterpris. .~cils m ?an. First, the Saturn local union is part of the

UAW international union. Traditionally, UAW contracts are negotiated company wide (i.e., a

national contract covers aUof GM’s facilities, another (ve~ similar) national contract covers Ford

workers, etc. Each plant negotiates a local supplement to the national contract covering work

rules specific to that operation. Saturn deviates from this arrangement by having its own

collective bargaining contract separate from the national agreement. However, the local union is

bound by the UAW constitution and the international negotiates with the Saturn Corporation and

assigns a representative to oversee the Saturn local. This international representative sits on the

Strategic Adviso~ Council at Saturn. The questio% therefore, does this structure allow the

UAW local at Saturn to seine as the fimctiomd equivalent of a Japanese enterprise union?

38

Some fmtures of the union’s role appear to be fictionally equivalent. First, union

members’ and officers’ long term economic and job security interests are tightly tied to the

suwival, growth and profitability of Saturn. By accepting a job at Saturn UAW members

originally gave up transfer rights to other GM jobs. The vast majority, therefore, hope to

complete their careers at Saturn. The compensation system at Saturn places a minimum often

percent of union member’s pay at nslq and provides a reward-sharing component that can (and

has to date) produced bonuses when the firm meets negotiated performance targets. -The temns of

the 1992 contract renewal made the risk portion of pay contingent on completion of training and

established a reward portion based on quality, output, and profitability goals that produced

bonuses of $2,600 in 1992, $3,000 in 1993,$6 p‘O in 1994 and $10,000 in both 1995 and 1996.

Michael Bennett, the president of Satu cal w Inception through 1995

also articulated a view of the union that emboc .ile esst .;e of the enterprise union perspective,

namely, that long term security is a result of workers’ contributions to firm performance, not a

right conferred through contractual negotiations. He notes that Saturn’s model of worker

representation is based on the premise that long term employment security cannot be negotiated

independent of the economic performance of the firrq nor solely through collective bargaining

after all strategic decisions have been made by management. Rather, in his view employment

security can only be achieved over the long run by both contributing to the economic performance

of the firm and participating duectly in business planning and decision-making processes to insure

that worker interests are given appropriate consideration (Bennett 1988).

Yet the Saturn local is a part of the UAW international union and carries over many of the

traditions structural features, and political processes that have been paxt of the UAW-GM

39

relationship for years. Some of these operate to limit the enterprise focus of the local union. For

example, the decision over whether to invest in a second generation product (Mod 2 in Saturn

language) (Hw 1994) has been heavily influenced by GM-UAW interests. GM’s corporate

executives and UAW national union leaders had to balance the desire of the local union at Saturn

for expansion and fi.lrther investment against the interests of other local unions competing for.

GMs investment dollars and new job opportunities. The national union leaders took the position

that it made little sense to expand operations at Saturn given the excess capacity that existed in

other GM plants slated for closure as their product lines were retired. Eventually, the UAW

national leadership’s views prevailed.

Contract negotiations provided another forum where’ d and national practices and

norms have come into conflict over issues such as length of, )rkdr . . . ~e~ shift

premiums, absenteeism policies, the role of seniority, the elec.. .,J versu~ ‘int selection of module

advisors, and the election of union representatives to handle grievances. In each case, the debate

over whether to “allow” or “accept” the local union’s proposal to depafi born national patterns

and practices centered around whether this would erode indust~ standards the union had fought

hard to achieve and to enforce. Thus, the union at Saturn has deeper interests and roles in the

enterprise that need to be balanced against its historic principles and the shared interest in

maintaining high labor standards throughout the indus~.

The local union also experienced considerable internal political turmoil and debate over its

role in the partnership. The first president of the local was challenged by an opposition candidate

‘Bb

40

(and an opposition “caucus’”) in 1993 elections. Similar opposition arose prior to the 1996

elections, leading the incumbent caucus to separate the roles of MAC advisor and president which

resulted in its retaining both positions. Suwey data and our own observations at Saturn clearly

showed a growing restiveness among rank and file members at Saturn throughout the 1992-96

time period. Our focus group interviews as well as the union’s internal member to member survey.

data found that much of the concern reflected a feeling that while the union was doing a good job

of representing the memberships’ collective interests through the partnership, it was applying

insufficient resources to its more traditional role as a representative of individual workers with

particular concerns or grievances. As a result, the 1994 labor agreement provided for the election

of a set of union representatives who would be responsible for handling m ‘r grievances in a

fashion similar to the traditional grievance committee member role of other ,icar

(including the UAW). These data also showed that throughout this period - ..~abie m, vity of

the membership continued to support the partnership and believed it was preferable to the

traditional UAW-GM relationship, while they wanted oppofiunities for more individual

representation of their specific concerns. Thus, the union implemented this change to achieve a

better balance between its role as a management partner which enhanced collective representation,

and its ability to achieve individual representation of workers’ specific interests.

ConiXct and its Resolution

‘T’heUAW has a long history of “caucuses”, i.e., groups that organize a slate of candidates forelected office. The “Reuthex Caucus” was in control of the national union leadership positionsfrom the election of Walter Reuther in 1948 to well after his death in 1970 (Lichtcmstein, 1995).currently there is an “admhktmtl “on” caucus and a “peoples” caucus vying for control within thelocal union at Saturn.

41

As the above examples begin to suggest, Saturn has experienced significant

intraorganizational conflict over its history. Two features of these conflicts stand out. First, they

seldom took the form of bilateral, labor-versus-management conflicts. Instead, they often

involved multiple groups with different and shifliig coalitions reflecting both different horizontal

and vetiical interest groupings. Some, as noted above, involved differences between the local and

the national union. Some involve differences among managers and workers located in the Spring

Hill manufacturing operations versus engineers located in Troy, Michigan. Some reflected

differences across the three business units (plants) in Spring Hill. Second, neither these conflicts,

nor those that did follow more traditional labor and management interest group lines, were

resolved crisply through a single negotiated agreement or unilateral management decision

Instead, confiicts played out in a more extended set of discussions, often taking place in, ie

fomms until either a consensus emerged or a crisis forced a final decision. One example. .,s

was a 1992 incident to protest what the union and its members perceived to be a weakening of

management commitment to building and shipping high quality products. In this case the union

members wore black armbands in the plant to symbolize their mourning of the changed

management policy. Another occurred over whether to choose a GM plant as a supplier in the

face of evidence indicating an outside, non-union supplier could deliver the part with newer

technology at a more favorable price. Other sources of cofllct arose over difficulties

Encountered in resolving production problems identified by work teams that required design

changes subject to approval f?om engineers located in Michigu differences of view over how

much autonomy should be afforded the separate business units (plants), concerns over how to

42

evaluate the performance of module advisors (supervisors), and issues involving the terms of

employment governing contract employees working in the cafeteria.

One fascinating conflict arose when the union leadership felt the company was moving too

slow in responding to production problems identified by rank and file workers. To document the

magnitude of these problems, the union conducted a sumey of its members and presented

management with a list of over 1,000 problems awaiting management attention. The presentation

was made at an afternoon meeting in a large bay of the assembly plant with over two hundred

workers and managers listening as workers, union representatives, and some of their management

allies summarized the list of problems for Saturn’s senior executives. As a result, a new problem-

solving initiative was put in place and more engineers from Michigan were relocated to the Spring

Hill facility to work on the accumulated list of problems.

Leadership Succession

Managing in a fashion that suppofis and sustains the type of partnership described above

requires a leadership philosophy and style that is not ofien found or rewarded within American

management. Saturn’s president from 1986 to 1995, Richard (Skip) Lefauve described his

approach to management as foUows:

We walk the line between providing inde~naknce and autonomy and havinginterdependence recognized and honored In this way, management and union balance oneanother, and the partnership bm”ngsto bear di~erent neeak for control. i’le union is in theroom, not talked to about the decision afterwards The process is inclusion, not consultation.Real value comesfiom shared ownership of the decision itseiJ which aIsopr&ces betteroutcomes.

By 1996 both LeFauve, the the top management leader, and Benne% the top union leader, who

developed and sustained the partnership principles through Saturn’s first decade were succeeded

43

bynew leaders. We botisuccessors espouse theprinciples of thekpredecessors, astillbe

noted below, they also face conflicting pressures horn other executives within GM and the

leadership of the international UAW to finction in more traditional ways. Whether the

partnership principles sutive this succession is uncetiain at this stage in the organization’s

history.

External Relations

Propositions 9 and 10 suggest that firms designed around stakeholder principles would

experience considerable opposition and resistance ti-omboth traditional business and labor groups,

albeit for ve~ different reasons. In fact, Saturn seems to be leading a rather schizophrenic

existence in its external relations. On the one hand, its success in the marketplace and the

receptivity of consumers and the general public to the “A Different Kind of Company, a Different

Kind of Car” slogan it used in its advertising has made it vexydiflicult for critics within GM to

publicly criticize Saturn. Business Week (1992: 86) put it as follows:

Saturn’s sudden blast up the sales charts is heafiening for it troubled parent. ...As foreignrivals continue to flood the market with new models, Saturn is meeting these head-on.Almost ovemigh~ Saturn has become the highest quality American-made brand, with asfew defects as Hondas and Nissans. It’s stunningly successful at satis@ing the customer,trailing only Lexus and Infiniti, according to researcher J.D. Power& Associates.So the auto maker clearly has a winner on its hands. Now the question is: WN GM knowwhat to do with it?

On the other hant there is significant resistance within GM to Saturn. The same Business Week

article cited above contained the fol!owing comments on the reaction of executives within GM’s

Chevrolet unit:

Call it a case of bad sibling nvahy. While wunderkind Saturn Corp. basks in the limelightwith its hot-selliig new cars, Chevrolet has been shunted aside... .“All that money that

44

went to Saturn during the past halfdozen years and the other GM divisions left Chevroletnaked.” (says a Chevrolet dealer) (p.90).

These views also limit the transfer of learning and innovations from Saturn to other parts of GM

even though one of Saturn’s stated objectives was to seine as a laboratory for experimentation

and learning for the corporation. An anecdote illustrates this point. At a briefing of senior GM

executives on the results of the network analysis we discussed how to get others in the company

to learn from these data. Two comments illustrate the essence of the problem better than any

others. One executive said:

It’s ironic but utiortunately true that Ford has learned more from Saturn than we have inGM.

Another said:

Is there some way you can present these data in a more generic form so that our managerswon’t know they are from Saturn? Once they know you’re talking about Satu~ they’lljust tune you out and say, ‘they’re different.’

Resistance is equally strong within the UAW international leadership. As noted earlier,

Saturn is identified with the leadership of Donald Ephliu the UAW vice president who led the

effort to get GM to invest in this project and led the campaign to get the UAW Executive Board

to approve the controversial provisions of the initial Saturn labor agreement. Ephlin subsequently

retired from the UAW in part because of differences in philosophy with other top UAW leaders.

The current president of the UAW has been critical of Saturn and opposes adopting the Saturn

contract to other plants. This is one reason why GM decided to build the second generation

Saturn in another GM plant rather than approve the request from Spring Hill to expand and build

the new model there. The UAW international leaders argued against expanding the Spring Hill

facility and favored an alternative proposal to build vehicles under the Saturn nameplate in another

45

plant that was scheduled to have excess capacity. The company and union decided on this

alternative and subsequently negotiated a local contract for this facility that embodies some of the

teamwork principles found at Saturn (and a considerable number of other UAW local agreements)

but kept the local agreement under the provisions of the national contract. The effect is to further

isolate the Saturn local union in Spring Hill and to limit the diffision of Saturn’s version of a

labor-management partnership.

Saturn’s visibility also makes it subject to plaudits and criticisms horn external sources.

For example, the Work in Anerica Institute, a national non-profit organization devoted to

promoting new models of labor-management cooperation and work innovatio~ has sponsored an

on-going stream of study tours to Saturn for national and international groups. On the other

Saturn was consciously not invited to testi~ before a national commission charged with the

tasKo~updating labor law because neither the national level labor nor the employer leaders who

were consulted on which organizations should testify wanted to showcase Saturn. It was too

controversial within both labor and business circles.

Finally, although there have not yet been serious legal challenges (apati horn the initial

challenge to the pre-hire union recognition agreement) to Saturn’s labor-management practices,

some of them clearly are inconsistent with the National Labor Relations Act. For example, the

deep involvement of workers and union officers in managerial roles is inconsistent with the

doctrine that there is to be a clear line of demarcation between bargaining unit members (workers)

and supervisors (managers). Extending voluntary recognition to the union before any employees

were hired i~~enerally illegal under labor law, even though the National Right to Work

Committee Iost’this argument in this particular case.

46

These are only isolated examples of the broader reality: the Saturn pafinership is the most

controversial innovation in labor-management relations and organizational governance found in

America today. It has ardent supporters and vigorous critics. It challenges deeply ingrained

ideological principles, traditions, and legal doctrines. It therefore serves as a symbol for a

“diffkrent kind of company with all the advantages and risks attendant to such a position.

Saturn’s ultimate fate probably depends less on its objective performance than on how these

external and internal political dynamics are managed and p!ay out (Hancke and Rubinstei~ 1995).

But the lesson we draw fiorn this highly charged debate is that stakeholder firms that embody

similar features to Saturn’s that increase employee influence and involve collective representation

will evoke similar intense debates. For these organizational forms to be widely adopted and to

.hlvlve changes in both corporate and labor law and considerable shifi in power and

.Jeoiug in society.

Discussion

Saturn is an evolving organization in which employees collectively have taken on many of

the characteristics of critical stakeholders and the organization’s structure and governance process

were designed in ways necessary for a stakeholder firm to succeed. The company can only meet

the objectives of its shareholders if employees contribute their knowledge to managing and

improving the production process. Our analysis of the co-management process documents at least

one aspect of the union-management partnership that contributes to the petiormance of the

enterprise. However, the long term sumival of this type organizational form is far from certain. It

remains vulnerable to several external threats in the American environment: (1) ideological

opposition to the expanded role of the union in decision-making and organizational governance,

47

and (2) a legal environment that discourages employees and their representatives from

participating in the management and governance processes of the firm.

In addition to these external threats, Saturn could eventually fail if it does not produce

sufficient financial returns to GM and/or it may lose the autonomy from the international union

and national labor contract that our analysis suggests is critical to its sumival as a stakeholder

form of organization. Continuity of support through successive CEOS and international union

pre&dents and officers is fm from guaranteed. Within Satu~ the failure to manage conflicts

effectively could also threaten the viability of the patinership by weakening employee willingness

to contribute their discretiona~ efforts to the firm.

AI° V@ our analysis focused on employees as stakeholders, Saturn’s design also

incorpori ,d det “rstakeholders as well. For example, the company’s marketing

strategy, . . .~red m s advertising slog~ “A Different Kind of Company, a DiiTerent Kind of

Car,” places tremendous emphasis on achieving and maintaining high levels of customer

satisfaction and loyalty. Thus Saturn carries over the same partnership principles that govern

employee relations into the relationships with its retailers. The trust and commitment embedded

in the retailer-producer partnership is essential for retailers to make another key put of the

company’s marketing strategy worlq that is its fixed price, “no haggle” sales practice. To avoid

the temptation of an individual dealer to defect from this strategy requires overcoming a classic

prisoner’s dilemma bargaining problem that can only be resolved and sustained by structuring the

incentives and the culture of the relationship in ways that allow both the manufacturer and the

retailers to prosper. Similarly, Saturn has moved toward single source suppliers linked to the

company with long term contracts and thus suppliers become a more critical resource to the

48

company than in the more traditional settings where the company purchases parts from multiple

competing vendors. Thus, the stakeholder principles and propositions developed here appear to

apply these relationships as well.

Analysis of the fill range of stakeholders embedded in this organization lies well beyond

the scope of our research. Yet by focusing on employees as stakeholders in this case we have.

derived a number of generic propositions for a stakeholder model of the firm that might apply to

other stakeholders as well. We encourage their examination and testing by other organizational

researchers.

It is not clear to us that Saturn or other organizations that attempt to adopt this

organizational form can - +ve in the present environment. Yet, regardless of Saturn’s fbture,

we expect to see increa Jmbe ltions adopt some or all of the features seen here

as human capital and m .lon-hr lcial assets rise in importance in organizations and begin to

challenge the supremacy of finance capital in shaping organizational strategies, designs, and

governance arrangements. If this is true, organizational theorists and practitioners will face a tall

order in rethinking basic concepts and practices and engaging the debates that lie ahead. Yet if

the social critiques of the American corporation and employment relationship outlined at the start

of this paper are valid, this tall order needs to be addressed both by the analytical research

community and by the policy makers and organizational leaders who will shape Iiture

organizations. We hope this paper sparks debate within the organizational theory research

community over these issues so that if and when the policy and organization! debates begin in

earnest, we will have something substantive to contribute.

IWWNCES

Aoki, hkahiko.

1988 Information. Incentives. and Banzainkw in the Japanese Economy, New York: CambridgeUniversity Press.

Ao& Masahiko.1990 “Toward an Economic Model of the Japanese Firm”, Journal of Economic Literature,Volume 28 (l): 1-27.

Bemetg Michael.1988 “New Roles for Unions in a Team Environrnen~” Manuscript, Work in AmericaProductivity Forum.

Berle, Adolph and Means, Gardiner.1932 Private ProDextvand the Modem Corooratio~ New York: MacMilIan.

Blair, Margaret.1995 OwnershiD and Control, Washir n, D.C.: The Brookings Institution.

Bla.si,Joseph R. and Douglas Kruse. “1991 The New Owners. Champai~ Ver .- .Ab

Bowma~ Edward and Michael Useem.1995 “The Anomalies of Corporate Governance,” in Edward Bowman and Bruce Kogut (eds.)Redesimirw the Firm. New York: Otiord University Press, 21-48.

Business Week.1992 “Saturn: GM Finally Has a Real Whner But Success Brings a Fresh Batch of Problems,”Auwst 17,86-91.

Chandler, AMed Dupont.1977 The %ible Hand: The Managerial Revolution in American Business, Cambridge, MA:Belknap Press.

Child, John.1972 “Organizational Stmcture, Environment, and Petiormance: The Role of Strategic Choice,”sociolo~ (6) 1, Janutuy, 1-22.

Chinoy, E.1955 Automobile Workers and the American Dream Garden City: California Press.

Cease, RH

50

1937 “The Nature of the F-” in O.E. Wdliamson and S.G. Winter (eds.), The Nature of theFirm: Orirnns. Evolution. and Development, New York: Otiord University Press.

Cohen, Susan G. and Diane E. Bailey1996 “Trends in Team EffectivenessResearch: From the Shop Floor to the Executive Suite,”unpublished paper, Center for Effective Organizations, University of Southern California.

Commission on the Future of Worker-Management Relations.1994a Fact Finding Repo~, U.S. Departments of Commerce and Labor, Washingto~ D.C.

.Commission on the Future of Worker-Management Relations.1994b Renort and Recommendations, U.S. Departments of Commerce and Labor, Washingto~D.C.

Cutcher-Gershetield, Joel.1991 “The Impact on Economic Performance of a Transformation in Workplace Relations,”Industrial and Labor Relations Review, Vol. 44, Number 2, (January) 241-260.

Donaldso~ Thomas, and Lee E. Preston.1995 The Stakeholder Theory of the Corporation: ~ ~~ts,E\~~---e, ~- -’Implications,”Academv of Mana~ement Review, Vol. 20, Numbe .91.

Dorfin~ Joseph.1949 The Economic Mind in American Civilization, New York <iking Press,

DiMaggio, Paul J. and Walter W. Powell.1983 “The Iron Cage Revisited: Institutional Isomorphism and Collective Rationality inOrganizational Fields,” American Sociological Review, Volume 48, April: 147-160.

DiMaggio, Paul J.1995 “Comments on ‘What Theory is Not’,” Administrative Science Outierlv, 40, September:391-397.

Drucker, Peter.1980 Manatimz in Turbulent Times New York: Harper and Row.

The Economi@.1996 “Stakeholder Capitalk~” August 10,23-25.

Ev~ Wlhm.1978 Orsmnization Theow Stmctures. Systems. and Environments. New York: Wiley.

FOL Alan.1974 Bevond Contract: Work. Trust. and Authoritv Relations. London: MacMillan.

51

Free- RE.1984 Stratem“cManatzement: A Stakeholder Approach Boston, MA: Pitman.

Freemtq R E. and Reed, D.1983 “Stockholders and stakeholders: Anew perspective on corporate governance,” in C.Huizinga (cd.), Corpo rate Govemanc~. Los Angeles: University Press, 89-106.

Free- Richard B., and James L. Medoff.1984 What do Unions Do?, New York: Basic Books.

Friedmq Milton.1958 capitalism and Freedow Chicago: University of Chicago Press.

Friedm~ Milton.1970 “The Social Responsibility of Business is to Increase its Profits,” The New York TimesMamtzine, September 13,

Galbrait~ John Kenneth.1967 The New Industrial State, Boston, Houghton MM1.in.

Gallup Organization.1995 Gallup Poll on Confidence in Institutions, PrincetoL NJ.

Glaser, Barney, and Anselm Strauss.1967 The Discovery of Grounded Theory: Stratesziesfor C)ualitativeResearch, Chicago, IL:Aldine Publishing Co.

Giddens, Anthony.1976 New Rules of the Sociolotzicd Method. New York: Basic Books.

Hancke, Bob, and Saul Rubinstein.1995 “Limits to Innovation in Work Organization?”, in Ake Sandberg (cd.) EMchimZ Production,Vermont: Ashgate Publishing Co, 179-198.

Heckscher, Charles.1995 White Collar Blues : Manamm ent Lovaltv in an Age of Comorate Restructuring, New York:Basic Books.

Hutto~ Wdliam.1995 The Stale Weire m London: Jonathan Cape.

Gould, WMarn~1994 Agenda for Refo~ Cambridge, MA MIT Press.

●

52

Je~ Karen A.1995 “A multimethod examination of the benefits and detriments of intragroup conflict,”Administrative Science Ouarterlv=(40) 256-82.

JenseL Michael C.1989 “Eclipse of the Public Corporatio~” Hamard Business Review, 89 (September-October, 61-74.

Katq Harry. “1985 ShifiimzGears: Chantinsz Labor Relations in the US Auto Industrv, Cambridge, MA: MITPress.

Koch~ Thomas.1995 “Using the Dunlop Report to Achieve Mutual Gains,” Industrial Relations, Vol. 34, Number3,350-366.

Kocharg Thomas, Harry Ka@ and Robert McKersie.1986 The Transformation of American Industrial Relations, New York: Basic J? s.

Kocha~ Thomas, Robert B. McKersie, and Peter Cappelli.1984 “Strategic Choice and Industrial Relations Theory,” Industrial Relations 2: ‘9.

Koch~ Thomas, and Paul Osterman.1994 The Mutual Gabs Entemrise: Forgimz a Wkmhz Partnership AmonszLabor. Mana~ement.and Government, BostoL NW Hamard Business School Press.

Kotter, J., and J. Heskett.1992 Comorate Culture and Performance, New York: Free Press.

Landers, Renee, James Rebitzer, and Lowell Taylor.1996 “Rat Race Redux: Adverse Selection in the Determination of Work Hours,” AmericanEconomic Review, Volume 86, Number 3, June: 329-348.

Lawler, E.1986 Hhzh Involvement Mana~ement, San Francisco, CA: Jossey-Bass.

Lichtenste~ Nelson.1995 The Most Darmerous Man in Detroit, New York: Basic Books.

M.i]~O~ Paul and Roberts, John.1992 Economics. Or~animations.and Manatzemen~. Englewood Cliflk, NJ: Prentice Hall.

Mills, D.Q.

53

1978 “Flawed VktOV in Labor Law Refo~” Harvard Business Review. Voume 57. Number 3,92-102.

McNulty, Paul.1980 The Orhzins and Development of Labor Economics, Cambridge, MA: MIT Press.

O’Toole, Jack.1996 Fonmn“ g the Future: Lessons from the Saturn Coloration. Cambridge, MA: Blachell.

Parker, Mike, and lane Slaughter.1988 Choo in id :~t, Labor Notes, BostoL MA: South EndPress.

Pfeffer, Jefiey and Gerald Salancik.1978 The External Control of Organizations: A Resource Deuendencv Pers~ective. New York:Harper and Row.

Reich, Robert.1996 “The New Corporate Citizenship,” U.S. Depatiment of Labor,l$%shingto~ D.C., Apri] 16.

Roe, Mark J.1994 Strong Manwwrs. Weak Owners, Princetow NJ: Princeton University Press.

Roy, Donald.1953 “Worker Satisfaction and Social Rewards in Quota Achievement,” American SociologicalReview, Vol. 18, October.

Rubinste.~ Saul A1996 Saturn. The GMIUAW Partnershb: The Inmact of Co-mana~ement and Joint Governanceon Firm and Local Union Performance, Ph.D. Dissertation Massachusetts Institute ofTechnology.

Rubinste@ St@ Michael Benn~ and Thomas Kochan.1993 “The Saturn Partnership: Co-Management and the Reinvention of the Local UnioR” inBruce Kaufinan and Morris Kleiner (eds.) Emplovee Representation: Alternatives and FutureDirection$ hfadiso~ WI: IRRA Press, 339-370.

Schek E.1983 “Role of the Founder in Creating Organizational Cukure,” Orm.nization Dynamics Summer,13-28.

Schmidt Stur@ and ThOIIMSKochan.

54

1972 “Conflict: Toward Conceptual Clarity,” Administrative Science Ouarterlv, September: 3s9.

370.

Sel.znic~ Philip.1996 “Institutionalism: ‘Old’ and ‘New,’ Administrative ScienceOuarterlv,41, 270-77.

Slichter, Sumner, James J. I+kaly,and E. Robert Livemash.1960 The Impact of Collective Bar~ainina on Manatzemen~ Washington D,C.: The BrookingsInstitution.

Stdord Research Institute.1963 Internal Memorandum.

Stem Robert N. and Stephen R. Barley.1996 “Organizations and Social Systems: Organization Theory’s Neglected Mandate,”Administrative Science Ouar&eriy~41, 146-62.

SuttoL Robert I. and Barry M. Staw.1995 “What Theo~ is Not,” Administrative Science Ouarterly, 40, September: 371-384.

Thurow, Lester C.1996 The Future of Caoitalisn New York: W. Morrow& Co.

Usee~ Michael.1996 Investor Capitalism. New York: Basic Books.

Waho~ Richard.1985 “From Control to Commitment in the Workplace,” =_Business Review, March-April:77-84.

WaIto~ Richard E., and Robert B. McISersie.1965 ~ Behavioral Theorv of Labor Negotiations: An Analvsis of a Social Interaction Svste~ILR Press, MM- NY.

Waito~ Richar~ Joel Cutcher-Gershenfeld, and Robefi B. McKersie.1994 Straternc N~tI “ations: A Theow of Change in Labor-Manastement Relations, Bosto~ MA:Harvard BusineSSSchool Press.

Weiclq Karl.1983 The Social Psvcholow of Oripnizin& (2~ ediiion), Reading MA: Addison-Wesley.

Weiclq Karl.1995 “what llwo~ is Not Theorizing Is,” Administrative Science Ouarterlv. 40, September385-390.

55

Weiler, Paul.1990 Governirw the Worlmlace: The Future of Labor and Employment Law. Cambridge, MA:Harvard University Press.

Wever, Kirsten.1995 ~egot iatinszComDetitivenes~ Bosto~ MA: Harvard Business School Press.

56

Table 1

1982:

1983:

1985:

1986:

1987:

1988:

1989:

1990:

1991:

1992:

1993:

1994:

1995:

1996:

Partnership Evolution and Deve]oDmentGM Small Car Study

Joint GMKJAW “Committee of 99” Study World Class Manufacturing

Memorandum of Agreement on Saturn Corporation

Joint Supplier Selection BeginsFti UAW.Hiring

Approval of Initial $1.9 BNion Saturn Capitalization by GM Board of Directors

Addition of Jointly Selected UAW Module Advisors as Partners

Addition of Jointly Selected UAW Crew Coordinators and Staff Partners

First Car off the LineFwst Local Union General Election

Second Crew AddedFirst Member-to-Member SurveyAgreement Renewal ProcessWithdrawal of UAW International Representative From SAC

First Election of Team Leaders (Work Unit Counselors)Second Member-to-Member Suwey

Local Refmendum on PartnershipSecond Local Union Election

Third Member-to-Member Suxvey

Saturn Integrated into GMs Small Car GroupSkip LeFauve Promoted to Head Small Car GroupAgreement Renewal Process Results in Establishment of 14 Elected Crew Coordinators

Election of 14 Crew Coordiitors with Authorization to File GrievancesDon Hudler, former VP Marketing replacesLeFauve as Saturn President.

Mike Besmett Resigns as Local Union Presidenk Retaining Position as MAC AdvisorJoe Rypkowski Su-tis Bennq becoming the 2~ of Pr~ldent of UAW Local 1853Bennett and Rypkowski both Wm in Third Local Union ElectionGM Announces Saturn Expansion to Wilmington Assembly Plant

57

~● Treat people as a fixed asset. Provide opportunities for them to maximize theircontributions and vaIue to the organization. Provide extensive training and skill development to allemployees.● The Saturn organization will be based on groups which will attempt to identi~ and workcollaboratively toward common goals.● Saturn WiIloopenlyshare all information inchding financial data.● Decision making will be based on consensus through a series of formal joint labor-management committees, or Decision Rings. As a stakeholder in the operation of Saturn theUAW will participate in business decisions as a Ml Partner including site selection andconstruction process and product desi~ choice of technologies, supplier selection make-buydecisions, retail dealer sekctio~ pricing, business planning, training, business systemsdevelopment, budgeting, quality systems, productivity improvement, job desi~ new productdevelopment, remitment and hiring maintenance, and engineering.8 Self-managed teams or Work Units will be the basic building blocks of the organization.● Decision making authority will be located at the level of the organization where thenecessam knowledge resides, and where implementation takes place. Emphasis will be placed on

nit.:re will be a minimum ofjob classifications.

._. Jm will have a iointlv develomd and administered recruitment and selection Process,. . . .‘~nd work units will participate in hiring their own team members. Seniority will not be the basisfor selectioz and the primary recmiting pool will consist of active and laid off GMAJAWemployees.● The technical and social work organization will be integrated.● There will be fewer fhll time elected UAW Officials and fewer Labor Relations personnelresponsible for contract administration.● Saturn’s reward system will be designed to encourage everyone’s efforts toward thecommon goals of quality, co% timing and value to the customer.

Source: Saturn’s 1985 Collective Bargaining Agreement

.

‘9

9

58

Table 3

Conununications and Quality Improvement

High Quality Improvement Low Quality Improvement T-statistic

Oremall CentralityRep Advisor

Non-Rep Advisor.

Group CentralityRep Advisor

Non-Rep Advisor

Quality consmunioationsRep Ad;isor

Non-Rep Advisor

c J1 Ct traliqRep Advisr

Non-Rep Advisor

Group CentralityRep Advisor

Non-Rep Advisor

Group DensityRep Advisor

Non-Rep Advisor

21.5 15.8 2.117**(8.13) (6.19)19.2 17.9 0.508(5.13) (4.99)

8.7 5.4 2.751**(4.02) (2.11)4.4 4.1 0.478(1.45) (1.7)

4.1 1.7 2.789**(2.08) (1.41)1.5 1.9 0.844(1.32) (1.11)

Communications and First Time Quality

High ~Q Low FTQ T-statistic

18.62(6.98)14.47(6.04)

7.438(3.86)4.647(1.41)

0.102(0.036)0.091

11.43 2.897***(5.4)15 0.232(5.79)

6.125 1.09(2.87)3.727 1.505(1.67)

0.059 4.006**+(0.024)0.104 1.025

(0.031) (0.033)Quality CoamunioatiolmRep Advisor 3.46 1.76 2.170**

(2.15) (1.44)Non-Rep Advisor 2.21 1.12 2.603’*

(1.13) (1.05)

N

32

32

31

N

32

32

32

31

* significant at the .10 level (standard deviations in parentheses)● * significant at the .05 level● ** significant at the .01 level

59

Vs

BS

PT

Figure 1. Communications Density Among Module Ad~ors

Saturn Site-WideT: 2.413**N: 114

Group: Non-Rep RepDensity: 23.3% 28.6%

Vehicle Systems Body SystemS@S) Powertrainm

T: 3.122*** 1.501 0.066N: 52 38 24

GrouIIx Non-Rep Rep Non-Rep Rep Non-Rep RepDensitj: 17.6%- 26.3% 23.7% 29.4%

IP .

#

■ um I

h●

m

9— ——~— . . ... ..m ..

9

Rep :“

“a

‘LL_l—L—

33.3% 33.0%

.. .

%

.—

4

9

—

P4i!3

..,

i!

ili!i!-:-gr

“i?

60